To what extent does job quality influence organizational commitment across countries?

(1)

To what extent does job quality influence organizational

commitment across countries?

Francesca Tunisi

Supervisor: Karen Modesta Olsen

Master thesis, MSc in Economics and Business Administration, Strategy and Management

NORWEGIAN SCHOOL OF ECONOMICS

This thesis was written as a part of the Master of Science in Economics and Business Administration at NHH. Please note that neither the institution nor the examiners are responsible − through the approval of this thesis − for the theories and methods used, or results and conclusions drawn in this work.

Norwegian School of Economics Bergen, Autumn 2019

(2)

Acknowledges

I would like to start by thanking my supervisor Karen Modesta Olsen, for her help and availability throughout the process of drafting my research thesis.

Further, I would like to thank the FOCUS-program for the great opportunity of collaboration and for the support during the research process.

Lastly, I also want to thank my family for support and care along the way of this course of study.

(3)

Abstract

Organizational commitment has shown strong and consistent relationship with positive outcomes for companies such as a reduction of organizational withdrawal behaviors and an increase in individual efforts at work.

This research investigates the relationship between job quality and organizational commitment and the role of country in shaping the two variables and their relationship.

The hypotheses have been tested through the dataset of the 2015 International Social Survey Programme on work orientations. Two countries have been taken into consideration for making the comparison: Norway and France. They exhibit differences according to the variety of capitalism theory’s clusters.

The statistical analysis shows that a positive and significant relationship exists between job quality and organizational commitment. In addition, differences among Norway and France are observed in the mean level of both constructs. However, no significant evidence has been found to support a moderating role of country in the relationship between job quality and organizational commitment. Results of the study suggest that the relationship between them is the same regardless the country.

Overall, this research has proven that institutional regimes differences account for the higher values of job quality and organizational commitment in Norway compared to France. However, the country element does not influence the strength and direction of the relationship.

(4)

List of Figures

Figure 1 - Research model 18

List of Tables

Table 1 - Total variance explained 27

Table 2 - Rotated component matrix 27

Table 3 - Mean and Standard Deviations in Norway 32

Table 4 - Correlations in Norway 33

Table 5 - Mean and Standard Deviations in France 34

Table 6 - Correlations in France 35

Table 7 - Regression analysis - Organizational commitment and job quality 36

Table 8 - Regression analysis - Extrinsic factor 37

Table 9 - Regression analysis - Intrinsic factor 38

Table 10 - Mean and standard deviation of organizational commitment 39 Table 11 - Regression analysis - Organizational commitment 39

Table 12 - Regression analysis - Moderated model 41

Table 13 - R values for the models 42

Table 14 - Regression analysis - Relationship in Norway 43 Table 15 - Regression analysis - Relationship in France 43

(7)

1 Introduction

1.1 Background

The concept of organizational commitment represents a topic of long-standing interests in the study of work behaviours. Commitment embodies attachment and loyalty to an organization. It has shown positive relationship with other relevant work and attitudinal constructs such as job satisfaction and it is widely related to positive outcomes for organizations (Meyer, Stanley, Herscovitch, & Topolnytsky, 2002). High levels of organizational commitment directly affect turnover, turnover intentions, absenteeism, job performance, organizational citizen behaviour and employee health and well-being.

High turnover rates destroy resources for companies, they are costly and reduce the profitability. Indeed, a new employee will require trainings and it will take time to be highly productive. Moreover, losing an employee can be seen as an investment waste and can stand for leaving knowledge and skills to the competition.

Absenteeism is the habitual non-presence at work of an employee, and it is related to organizational commitment (Steers, 1977). Employees who are committed to an organization are more likely to positively contribute to the firm’s performance and to come to work with higher desire.

Companies can benefit from committed employees because they engage in behaviours that generate positive outcomes. Examples of such behaviours are extra efforts, help to co-workers and longer working hours.

Implications of organizational commitment for individuals are less work-related stress and higher self-esteem (Fornes & Rocco, 2004). Committed employees show less

(8)

emotional exhaustions even though they work longer and harder than those who are not committed.

So far, attention has been placed on the positive outcomes of organizational commitment. However, in order to create a positive and supportive environment it is important to understand what drives it. By knowing what conditions foster organizational commitment and which are the elements promoting it, organizations will be in a better position to act and react to changes more effectively.

Antecedents of organizational commitment have been traced in personal characteristics, work experiences and job-related characteristics (Meyer & Allen, 1991;

Mercurio, 2015; Steers, 1977). In particular, this study will focus on the relationship between the concept of job quality and organizational commitment. Job quality is a multidimensional concept comprising job features that have a relationship with health, physical and psychological well-being.

The analysis is conducted through the data of the 2015 International Social Survey Programme (ISSP) Module on work orientations. The ISSP is a self-funding association that conducts annual survey on social sciences topics. Surveys are replicated through years and across-countries in order to encourage comparisons. In particular, the survey on work orientations has been replicated four times. Such dataset supports a cross-national comparison of the relationship being analysed in this study.

The concept of organizational commitment has raised interest on the degree to which cross-national differences exist (Hattrup, Mueller, & Aguirre, 2008; Lincoln & Kalleberg, 1985; Mowday, Steers, & Porter, 1979). Norway and France will be compared through this research. Indeed, significant differences have been highlighted between these two countries in the light of the varieties of capitalims theory (Amable, 2003; Gallie, 2011).

Social democratic countries such as Norway, are characterized by more favourable

(9)

employment rights and a high degree of partecipation of the organizaed labor in organizations. These elements enhance job-related characteristics as job secutiy and working conditions, factors which influence job quality and organizational commitment.

This research aims to investigate whether job quality influences organziational commitment and to what extent countries differences affect the two (Hofstede &

Hofstede, 2005) concepts and their relationship. Understanding the causes and consequences that affect organizational commitment level may be of importance in institutional and organizational settings.

1.2 Research objective

The objective of this paper is to provide awareness on differences occurring across nations in the relationship between job quality and organizational commitment.

Previous recognized factors of job qualities influencing organizational commitment are used in the research.

The research question is:

To what extent does job quality influence organizational commitment across countries?

1.3 Structure

In order to respond to the research objective, this study will have the following structure. Firstly, a review of existing theories will be presented on organizational commitment and job quality. Based on these theories, a conceptual model and hypothesis will be developed. Secondly, the methods used to collect and analyze the data will be explained. The results will be then showed with a final discussion on the main findings and their possible contribution to future research. Finally, the conclusion is at the end of this thesis work.

(10)

2 Literature review

In the following section theories regarding the main research areas will be presented.

They have been thoroughly chosen in order to gain insights on the knowledge that has been reached so far.

During the first phase of research, I looked for papers that deal with organizational commitment in general without focusing on specific antecedents and relationships.

Building on what I discovered, then I refined my research and focused on the main topics in order to answer the research question.

2.1 Organizational commitment

Organizational commitment has been generally defined as the extent to which employees feel part and identify themselves with a particular organization (Bishop, Scott, & Burroughs, 2000). Mowday, Steers and Porter (1979) portray organizational commitment with three factors:

1) a strong belief in and acceptance of the organization’s goals and values;

2) a willingness to exert considerable effort on behalf of the organization; and 3) a strong desire to maintain membership in the organization.

Meyer and Allen (1991) conceptualized organizational commitment as a psychological state that groups three different themes. Their model, the three-component model, describes an employee’s relationship with the organization. The first component is the affective commitment and it reflects the idea of Bishop, Scott, and Burroughs (2000) and of Mowday, Steers, and Porter (1979). Indeed, Meyer and Allen (1991) define it as the “employee’s emotional attachment to, identification with, and involvement in the organization”.

(11)

The second component is the continuative commitment and it takes into consideration perceived costs. Commitment can be seen as the non-termination of an action because of the costs associated with it. In this sense, employees remain with the organization because they need to do so and not because they want to.

The last component of this definition is the normative commitment. In this view, there is a sense of obligation where employees remain with the organization just because they ought to do so.

The three-component model is considered a dominant model in organizational commitment research. Nevertheless, several other researchers have found it inconsistent with empirical findings. Solinger, van Olffen, and Roe (2008) support the idea that the concept of organizational commitment refers to an attitude towards the organization, while normative and continuative commitment are just attitudes regarding specific forms of behavior, that is remain or leave the organization. For this reason, authors have proposed to use just the first component of the model, affective commitment, when analyzing organizational commitment (Solinger, van Olffen, and Roe, 2008; Mercurio, 2015). Solinger, van Olffen, and Roe (2008) suggest defining organizational commitment as “affective attachment to an organization”.

For this reason, this last definition and the definitions of Bishop, Scott, and Burroughs (2000) and Mowday, Steers, and Porter (1979) will be used in this research.

Higher levels of organizational commitment are desirable for companies because it is correlated to a variety of positive outcomes. Generally, the effects and outcomes can be traced in lower levels of absenteeism and turnover, superior financial performances, improved production and improved overall performance (Fornes and Rocco, 2004;

Meyer, Stanley, Herscovitch, and Topolnytsky, 2002).

(12)

One of the most cited primary consequences of organizational commitment is turnover (Mercurio, 2015). Indeed, employee retention is a major concern for companies and by definition a committed employee is desirous of remaining within the organization and thus is less likely to leave. Different studies have found a negative correlation between commitment and turnover (Mowday, Steers, & Porter, 1979; Solinger, van Olffen, & Roe, 2008; Steers, 1977).

A less strong but still a relationship have been found between commitment and abstenteism. Higher levels of organizational commitment have been shown to lower levels of absenteism rate (Solinger, van Olffen, & Roe, 2008).

Employees with high organizational commitment also adopt behaviors that companies can benefit from: they are more willing to engage in extra role performance, they go above their simple role and responsibilities, they arrive early at work and try to provide all their knowledge and capabilities in trying to solve problems (Fornes & Rocco, 2004).

They also exhibit organizational citizenship behaviors, meaning discretionary behaviors and extra efforts that are not recognized by a formal evaluative system (Mercurio, 2015).

Organizational commitment is positive not only from a company’s perspective, but it can be beneficial also for employees. Indeed, employees with high levels of organizational commitment have lower levels of work-related stress and higher levels of self-esteem (Fornes & Rocco, 2004; Mercurio, 2015).

2.2 Antecedents of organizational commitment

Considering the fact that organizational commitment is a desirable outcome, it is important for companies to understand which are the actions or elements that cause commitment to occur.

(13)

Different studies categorize antecedents of organizational commitment in three clusters: personal characteristics, work experiences and job-related characteristics (Meyer & Allen, 1991; Mercurio, 2015; Steers, 1977).

Personal and demographic characteristics such as age, education, need for achievement have been showed to be linked to commitment but relations are not always strong and consistent (Meyer & Allen, 1991; Meyer, Stanley, Herscovitch, &

Topolnytsky, 2002).

Work experiences can be meant as group attitudes toward the organization, meeting expectations, personal importance in the organization, socialization and interpersonal relationship. These have been found to be positively correlated to organizational commitment (Steers, 1977; Mercurio, 2015).

Different job-related characteristics have been found to be positively related to organizational commitment. In a meta-analysis Fornes and Rocco (2004) identified five job characteristics as antecedents: congruency, whenever employee’s values and interests fit with the ones of the organization, the individual become more emotionally committed to the organization; having an interesting work means having a job that is challenging and rewarding; clarity of purpose, that is having clear information about intentions, ideas, goals and plan of an organization, it makes easier for employees to be informed and have a clear sense of direction; providing feedback, organizational commitment can be enhanced thanks to a continuous exchange of feedback about how an employee is performing and how he can improve his performance; autonomy, organizational commitment is stronger when employees have the chance to perform well on their job and at the same time manage how to do it.

(14)

2.2.1 Job quality

In a more comprehensive view, different indicators of the quality of work have been identified by the literature as influencing organizational commitment. Job quality can be defined as the extent to which a job has factors that enhance positive and beneficial outcomes for the employee, particularly psychological and physical well-being (Holman, 2013). Job quality is thus a broad concept that according to Kallenberg and Vaisey (2005) can be measured in two ways. The first one consists of evaluating job quality on a number of dimensions of work, while the second approach asks directly workers to grade their job. The latter method usually consists in asking the degree of job satisfaction. On the contrary, in order to evaluate job quality through the former approach, authors selected some job characteristics that are relevant for job quality estimation: economic benefits meant as earning and fringe benefits; non-economic benefits as autonomy and intrinsic rewards; job security and opportunities for advancement (Kellenberg and Vaisey, 2005; Holman, 2013). Such dimensions have been found to be positively correlated to organizational commitment (Jandaghi &

Bahrami, 2011; Fornes and Rocco, 2004; Steers, 1977; Meyer, Stanley, Herscovitch,

& Topolnytsky, 2002).

2.2.1.1 Job security

Job security can be defined as the feeling of having a proper job and the assurance of its continuance in future as well as the absence of threatening factors. Having no assurance of job can cause ill-health and job dissatisfaction (Dahl, Nesheim, & Olsen, 2009). Job security is linked with the uncertainty of work namely the possibility of losing the job, of wage cuts and of missed promotion opportunities.

(15)

2.2.1.2 Level of income

Pay is considered as a core dimension of quality of jobs. High level of income is positively considered by employees because it can be an indicator of status and freedom or autonomy (Dahl, Nesheim, & Olsen, 2009). Linked to wages are the topics of fairness and inequality. Indeed, a pay is considered fair when there is a clear connection with the work contribution to the company performance.

2.2.1.3 Advancement opportunity

Companies that provide constant training and development scheme increase the opportunities for employees to grow. This generates a positive consideration of their own work by employees.

2.2.1.4 Interesting work

Evaluating the work in a positive way is a central aspect of job characteristics because it can enhance motivation. Having an interesting work means that the employee judges the tasks as challenging and rewarding and allows him to use his knowledge and capabilities (Hackman & Oldham, 1976).

2.2.1.5 Job autonomy

Job autonomy can be found in an extensive literature regarding job redesign where the idea is to change work dimensions to enhance performances. One of the best- known works about job redesign is the one by Hackman and Oldham (1976). They developed a theory on job characteristics affecting employees’ motivation to perform effectively. Indeed, the model they presented is based on the relationship between job characteristics and individual reaction to the work. They identified five core dimensions among job characteristics that pursue favorable work outcomes: skill variety, task identity, task significance, feedback and autonomy. They defined autonomy as “the degree to which the job provides substantial freedom, independence, and discretion to

(16)

the individual in scheduling the work and in determining the procedures to be used in carrying it out” (Hackman and Oldham, 1976, p. 258).

2.3 Differences among countries

Up to this point, organizational commitment and job quality have been described at an individual and organizational level. However, the increasing globalization of businesses needs the generalizability of theory and constructs across countries.

Cross-country analysis can focus on two types of hypothesis: the first one addresses the effects of country on the level of a construct; the second one addresses the effects of country on the relationship between constructs (Hattrup, Mueller, & Aguirre, 2008).

Cross-national studies have found that institutional setting matters for work organization. Variety of Capitalism is a theory which has been particularly analyzed in this respect and it emphasizes the idea that institutional similarities and differences may produce distinctive outcomes for job quality and employment. The theory was first developed by Soskice and Hall (2001). They argued that different production regimes can take place whether differences occur across the financial system, the industrial relations system, the educational and training system, and the inter-company system.

Countries have been categorized according to how these systems interact among each other. In particular, they recognize two categories of regimes: liberal market economies (LBE) and coordinated market economies (CME). LBEs rely on market for coordination of the financial and industrial system while CMEs have higher degree of non-market coordination. In LBEs labor markets are less regulated and the financial system is guided by a short-term time horizon. The two regimes have been shown to have differences also in the skill systems. Indeed, liberal countries rely on more general education and skills whereas coordinated countries rely on specific skills and training.

This characteristic can have effects on different dimensions of job quality: autonomy,

(17)

opportunities for skill acquisition and job security. Coordinated market economies have comparatively higher wages and higher labor costs than liberal economies, thus they have to compensate this disadvantage through quality that is highly skilled workforce.

Specific skills are usually related to a particular occupation and firm meaning that the employee will probably have a long-term relationship with the company (Soskice, 1999).

Among countries, Soskice and Hall have identified the USA and the UK as best- representative for liberal market economies and the Nordic countries, Germany, Switzerland, the Netherlands, Belgium and Austria as coordinated market economies.

Starting from the categorization of Soskice and Hall, different authors have enlarged the number of cluster of countries with the aim of extending the analysis. Gallie (2011) added to coordinated and liberal market economies other three regimes: he split CMEs in two cluster Nordic countries (Denmark, Finland, Norway, Sweden) and Continental- coordinated countries (Germany, Belgium, Austria); State-coordinated countries (France, Italy, Greece, Spain, Portugal) and Transition countries (Estonia, Latvia, Poland, Hungary). Another way to cluster employment regimes has been developed by Amable (2003). He distinguishes between social democratic regimes (Nordic countries), continental regimes (Germany, France), liberal regimes (UK) and Southern Europe regimes (Spain, Greece).

According to the latter clustering, social democratic countries are characterized by a tight labor market because of favorable employment rights and a high degree of participation of the organized labor in the organizations’ decision-making process. This participation enhances job security, high employment levels and working conditions (e.g. wages and flexible hours).

(18)

Continental countries have instead organized labor that is only partly influential over organizations and government, thus its power to improve working conditions is weaker.

Social democratic regimes are thus more likely to have higher levels of job quality than continental regimes (Holman, 2013).

Moving to liberal regimes, these are characterized by little state interventions on working conditions because in this view, employment levels are best regulated by the market. Also, organized labor has little involvement in the organizations’ decision- making process, thus having just a marginal role in influencing working conditions. The low level of employment protection makes changing jobs more easily, thereby employers may invest less on training because of the employee possibility to leave the organization. Moreover, a lower skill levels may limit the creation of more complex jobs leading to jobs highly standardized with lower levels of discretion and therefore, lower pay levels.

Southern Europe regimes can be distinguished because of the low level of state intervention in the regulation of working conditions. The organized labor has just a partly influential role meaning that its ability to enhance employment condition is weak.

Moreover, investment in training and education is limited and thus, as in liberal regimes, this limit the design of complex jobs. In turn, this leads to low-quality working conditions and lower wage rates.

The institutional regimes theory and differences among countries in the job quality level, has been supported in different studies (Olsen, Kalleberg, & Nesheim, 2010;

Holman, 2013; Gallie, 2011). Olsen, Kalleberg and Nesheim found support of the varieties of capitalism approach beacause they found that job secutity, job autonomy and the quality of working conditions were higher in those countries belonging to the

(19)

coordinated market economies (Norway and West Germany) than in liberal market economies (Britain and US).

Regarding organizational commitment, differences across countries can emphasize it in different ways. Differences exists across countries that differs significantly in the dimension of the economy. Countries with a smaller economy show greater organizational commitment because of the important material interdependencies between generations that creates emotional attachment. Coutries with larger economies, instead, provide more job opportunities and therefore the emotional ties is weaker (Fischer & Mansell, 2009).

Moreover, as previously mentioned, coordinated market economies rely on highly skilled employees and compete on quality rather than on low costs. Specific education and skills make employees more linked to a specific occupation and organization.

While, in liberal market economies a lower skilled workforce makes the relationship between employees and employers characterized by a short-term perspective.

According to these differences, also work practices differs among countries. Indeed, coordinated and nordic countries have more “skilled-oriented” practices, while liberal economies such as the US and Canada have more “rule-oriented” practices. Skilled oriented practices have been found to lead to higher commitment levels (Dobbin &

Boychuk, 1999). However, there might be a tendency in social democratic and coordinated countries that might reduce the level of organizational commitment. As a matter of fact, a high level of organized labor involvement creates a strong collective identity resulting in contrasting feeling towards the organization (Hult & Svallfors, 2002).

On the second type of hypothesis, little literature exists on the moderating role of country in the relationship between job quality and organizational commitment.

(20)

Moderating a relationship means affecting the direction or strength of the relationship between the dependent and independent variables. In particular, when a moderator is found significant it can amplify or weaken the relationship between variables.

Some differences have been found in the relationship between job satisfaction, that, as previously mentioned, can be a proxy of job quality, and organizational commitment.

Lincoln and Kalleberg (1985) have found that the relationship was stronger among American employees than in the Japanese one. They suggested that the differences can be explained by the diverse work structure and work services. Other authors have suggested that there are differences in the relationship among individualistic and collectivistic countries (Hattrup, Mueller, & Aguirre, 2008). Goodman, Sabharwal and Chordiya (2017) also found differences in the relationship between job satisfaction and organizational commitment in the public sector of US and India.

2.4 Norway and France

Norway and France will be compared throughout this research. These two countries have different characteristics regarding the institutional setting and shows different employment conditions even though they are two European developed countries.

Indeed, a comparison between them may show which elements enhance job quality and organizational commitment.

Norway is considered a social democratic country where the organized labor is highly participant in the organizations and government decision making processes. This enhance job quality for employees. On the other hand, France is a continental country, where the participation of the organized labor is weaker thus it has less power to influence organizations and government to favorable working conditions.

(21)

2.5 Research model

The model that this research aims to test is a moderated one. In particular, it pursues to evaluate the impact of country in the relationship between job quality and organizational commitment.

Figure 1 - Research model

2.5.1 Hypothesis

As a result of the literature review, four hypotheses will be tested in this research.

The construct of job quality has been recognized as an antecedent of organizational commitment by Jandaghi & Bahrami (2011), Fornes and Rocco (2004), Steers (1977), and Meyer, Stanley, Herscovitch, & Topolnytsky (2002) thus the following is hypothesized:

Hp1: Job quality has a positive relationship with organizational commitment.

Successively, job quality has been found to be higher in social democratic countries than in continental country by Holman (2013). Also, organizational commitment is

Job Security

Level of Income

Advancement opportunities Interesting Work

Job Autonomy Job Quality

Organizational Commitment Country

(22)

higher in smaller economies than in larger ones. For these reasons, two hypotheses are formulated:

Hp2: Job quality has higher values in social democratic countries (Norway) than in continental countries (France).

Hp3: Organizational commitment is higher in Norway than in France.

Finally, based on the differences among countries in job quality and organizational commitment and on the works of Lincoln and Kalleberg (1985) and of Goodman, Sabharwal and Chordiya (2017), there is reason to believe that country has a moderating role in the relaltionship between job quality and organizational commitment. If country has a moderating role in the relationship, this means that among the two countries there would be a difference in the streght and in the direction of the relationship and thus one country would have a stronger link between the two variables.

Hp4: Country moderates the relationship between job quality and organizational commitment.

(23)

3 Method

The chapter on method includes a description of the research approach and of the research design. Moreover, there is an explanation of the data collection technique and the strategy that will be applied in order to analyzed them. Finally, there is a description of the data quality challenges.

3.1 Research design

A research can use either a deductive or inductive approach. In a deductive approach, a research starts with theory and theory-based hypothesis are developed in order to explain a causal relationship between variables. On the contrary, in an inductive approach, data are collected in order to explore a phenomenon with the aim to generate a new model (Saunders et al, 2016). The research approach used in this case is the deductive one since theories already presented on job quality and organizational commitment are the starting point of the research. Starting from the literature reviewed, I formulated a set of hypotheses and then I tested them.

Researches can have different purposes: explore, describe or explain. Exploratory studies aim to discover what is happening regarding a particular phenomenon and gain insights about it. Descriptive research wants to gain an accurate description of a particular event or situation. Explanatory research tries to establish causal relationship between variables (Saunders et al, 2016). This research aims to test hypotheses about the relationship between job quality and organizational commitment and thus it can be said that it has an explanatory purpose.

The hypotheses are tested using quantitative data. These types of data are numeric and are the result of any data collection technique that generates numerical data.

Quantitative and deductive researches are commonly associated with a survey

(24)

research strategy. Survey strategy allows the collection of standardized data in an economic way. This strategy does not allow to have in-depth understanding of the intentions of the respondents because of a limited number of questions. On the other hand, having standardized questions allows an easy comparison among answers (Saunders et al, 2016). The survey strategy has been used in this research and in particular the questionnaire is the data collection techniques employed.

3.2 Data collection

The data used in this research were those collected for the 2015 ISSP (International Social Survey Programme) Module on Work Orientations.

The ISSP is a self-funding association founded in 1984 with the aim of conducting annual survey on topics relevant to social sciences. Surveys are designed in order to be replicated to conduct cross-national and cross-time comparisons. Surveys also respond to new trends and development in social sciences by including new topics or developing new modules.

Work orientations is just one of the many modules conducted by the ISSP and the one of 2015 is the fourth module on this topic, previous waves were fielded in 1989, 1997 and 2005. This module mainly deals with topic such as employment arrangements, job characteristics, work outcome, work-life balance, attitude towards work.

In this module 37 countries were analysed, and respondents were aged 18 years and older. Data were collected through different method:

• Face-to-face interview: CAPI (Computer Assisted Personal Interview)

• Face-to-face interview: PAPI (Paper and Pencil Interview)

• Self-administered questionnaire: Paper

• Self-administered questionnaire: CASI (Computer Assisted Self-Interview)

(25)

• Self-administered questionnaire: CAWI (Computer Assisted Web Interview)

• Telephone interview

3.2.1 Sampling process

This research is analysing country-level data and therefore it would be impossible to collect information from the entire population. Sampling is thus necessary. Sampling procedures can be of two types: probability or non-probability. Probability samples is also called representative, indeed with this procedure the chance of each cases to be selected from the population is known. This means that it is possible to make statistical inferences about the characteristics of the population starting from the sample. On the contrary, with non-probability sample, the chance of each cases to be selected from the population is unknown. Therefore, it is not possible to statistically estimate the characteristics of the population (Saunders et al, 2016).

The ISSP declares that the sample procedure utilized was a probability sample with the following techniques:

- Simple Random Sample;

- Systematic Random Sample;

- Stratified Sample;

- Stratified Sample: proportional and disproportional;

- Multistage Sample.

3.2.2 The survey

The survey has been submitted in 37 countries. Questions are divided in multiple parts:

firstly, there are questions addressed to all the respondents on work centrality, values, work-life balance. Secondly, there are questions addressed only to those who are currently working for pay on job characteristics, subjective experiences, job

(26)

satisfaction, organizational commitment, job flexibility. Thirdly, there is a part for those not currently working. In this case, some questions regard past working experiences, reasons for not working, employability and job seeking activities. Finally, the last part of the survey regards optional questions on themes such as: recent work history, a self-assessment of economic situation, health.

Numerous variables on demographic data can be found in the survey: age, gender, education, hours worked weekly, type of organization (for-profit vs. non-profit and public vs. private), occupation, main employment status, family income and many others.

The sample for France and Norway includes respectively 4.500 and 4.400 interviewed.

The response rate has been 31,3% for France and 36,6% for Norway.

3.3 Measures

3.3.1 Organizational commitment

Based on the definition of Mowday, Steers and Porter (1979) described in the literature review, organizational commitment can be traced in the three elements of question 24 of the 2015 International Social Survey Programme (ISSP) on Work Orientation.

To what extent do you agree or disagree with each of the following statements? (1:

Strongly agree – 5: Strongly disagree):

a. I am willing to work harder than I have to in order to help the firm or organization I work for succeed.

b. I am proud to be working for my firm or organization.

c. I would turn down another job that offered quite a bit more pay in order to stay with this organization.

(27)

The job quality variables can be extracted from question 12 of the survey which is about job characteristics. In particular this part of the questionnaire is addressed to

“currently working” respondents.

For each of these statements about your (main) job, please tick one box to show how much you agree or disagree that it applies to your job. (1: Strongly agree – 5:

Strongly disagree) a. My job is secure b. My income is high

c. My opportunities for advancement are high d. My job is interesting

e. I can work independently 3.3.3 Control variables

Studies have found that age and gender have a positive relationship with organizational commitment (Mowday, Steers, & Porter, 1979). Thus, these two variables have been used as control variables. In addition, the type of organization (private or public) and the type of occupation have been used as control variables in order to isolate their effects from the effect of the main variables in the study.

3.4 Data Analysis

The research model has been tested using the Statistical Package for Social Sciences (SPSS).

Firstly, I performed the Cronbach’s Alpha analysis. Secondly, I performed a Principal Component Analysis on SPSS to understand the relationship among the items

(28)

composing the job quality index. Thirdly, I proceeded with different regression analyses and a two-step regression analysis to test the main hypotheses.

3.4.1 Cronbach’s Alpha

The Cronbach’s Alpha analysis has been performed for organizational commitment.

Cronbach’s Alpha is a common measure of internal consistency. In particular, it measures how much a set of items are related as a group (Bonett & Wright, 2015). Its value can range from 0 to 1 and most statisticians agree upon the fact that values above 0.7 are required in order to ensure that the questions considered are actually measuring the same construct.

The variable of organizational commitment is an index containing three different aspects:

a. The willingness to work harder to help the firm to succeed.

b. Being proud of working for the firm.

c. The willingness to turn down another job that offered a bit more pay to stay with the firm.

The Cronbach’s Alpha of these three items resulted in .70 that is the minimum acceptable in order to ensure that the questions combined are measuring the same construct.

3.4.2 Factor analysis

In order to establish unidimensionality, an exploratory factor analysis has been performed for job quality. The test has been conducted using the Principal Component Analysis (PCA) with the VARIMAX rotation on SPSS.

Before performing a Principal Component Analysis, it is recommended to assess the Barlett’s test of sphericity and the Kaiser-Meyer-Olkin (KMO) measure of sampling

(29)

adequacy. The Principal Component Analysis can take place when the the Barlett’s test of sphericity is significant (p<.05) and the KMO index is higher than .6 (Denis, 2018).

The number of factors to include in subsequent analysis can be determined through the Eigenvalues. Whether the factors have an Eigenvalue above 1, they can be included in the model. A similar result can be obtained through the cumulative percentage of variance extracted by the factors. Usually, the factors to be considered should account for around 80% of the variance (Denis, 2018).

The components identified by the analysis are then rotated through the VARIMAX rotation.

A check on multicollinearity has been performed on the result of the factor analysis.

The method used is the Variance Inflation Factors (VIF) in SPSS. Usually, VIF values above 5 or 10 as well as tolerance values under 0.1 indicate the presence of multicollinearity (Saunders et al., 2016).

As regarding the job quality index, a factor analysis has been implemented. The variable considered were all the job characteristics: job security, high income, advancement opportunities, interesting work and job autonomy.

The KMO measure of the sample is .633 and, hence, acceptable since greater than .60. The Bartlett’s test of sphericity also indicates that a factor analysis can be useful since it is highly significant (p ≤ .001).

(30)

Table 1 - Total variance explained

Component Initial

Eigenvalues Extraction Sums of

Squared Loadings Rotation

Sums of Squared Loadings

Total % of

Variance Cumulative

% Total % of

Variance Cumulative

% Total % of

Variance Cumulative

% 1 1,885 37,697 37,697 1,885 37,697 37,697 1,569 31,371 31,371 2 1,075 21,496 59,193 1,075 21,496 59,193 1,391 27,822 59,193 3 0,843 16,852 76,045

4 0,659 13,177 89,222

5 0,539 10,778 100,000

Extraction method: principal component analysis.

The Eigenvalues for the first two factors are greater than 1. Moreover, these components account for approximately 60% of the total variance, suggesting that the scale items are unidimensional.

Additionally, the VARIMAX rotated matrix has been performed.

Table 2 - Rotated component matrix

Component

1 2

Job Security 0,655 -0,100

High Income 0,761 0,174

Advancement Opportunities 0,716 0,253

Interesting Work 0,216 0,758

Job Autonomy -0,020 0,844

Extraction method: principal component analysis.

Rotation method: Varimax with Kaiser Normalization.

a. Rotation converged in 3 iterations.

The first component is linked with job security, high income and advancement opportunities, while the second one with interesting job and job autonomy. The two components can be seen as defining two different aspects of job quality: the first one is linked with extrinsic aspects of it, while the second one is related to job task characteristics and thus can be defined as an intrinsic component of job quality.

(31)

These two components, extrinsic and intrinsic factors, will be used in the subsequent analysis as representing job quality.

Among these two components, a multicollinearity check has been performed. The Variance Inflation Factor resulted around 1 meaning that the two factors do not correlate too highly.

3.4.3 Regression analysis

The research model is tested using multiple regression analysis on SPSS statistics. A regression analysis is a statistical method used to verify the relationship between a dependent variable and one or more independent variables and most importantly to examine the strength of impact of a number of independent variables on the dependent one.

The basic formula for a single linear regression with only one independent variable is:

𝛾 = 𝛽_%+ 𝛽_'𝜒_'+ 𝜀

Where 𝛾 is the dependent variable, 𝛽_% is the constant, 𝛽_' is the coefficient of the independent variable 𝜒_', and 𝜀 is the error term. The bivariate model can include a single independent variable and leave potential others unobserved. On the contrary, a multiple regression analysis can include an indefinite number of independent variables.

Its general equation is:

𝛾 = 𝛽_%+ 𝛽_'𝜒_'+ 𝛽_*𝜒_*+ ⋯ + 𝛽_,𝜒_, 𝜀

In the multiple regression analysis, each beta coefficient has to be interpreted as the effect of the specific independent variable on the dependent variable when all the other independent variables are hold constant.

(32)

To measure the degree of variation explained by a model, the adjusted R² is used.

Indeed, it measures the explanatory power of the model corrected for the number of predictors included.

3.4.4 Assumptions in regression analysis

When evaluating a regression analysis some checks are needed. First, the relationship between the dependent variable and the independent variables needs to be linear. In order to check linearity, a scatter plot has been created to visually inspect the relationship. The relationship between the variables is linear.

Second, the independent variables need to be not correlated because collinearity between variables would make it difficult to evaluate the effect of each one on the dependent variable. In order to check for collinearity, the Variance Inflation Factor can be implemented. Large VIF value indicates high collinearity (Saunders, Lewis, &

Thornhill, 2016).

3.5 Reliability and validity

Reliability and validity are central when assessing the quality of research. Reliability refers to “the extent to which the data collection techniques will yield consistent findings” (Saunders et al, 2016). A distinction has to be made between internal and external reliability. Internal reliability is about ensuring consistency during a research project. This may be achieved, where possible, through the comparison and discussion among more than one researcher on the methods to use to conduct interviews and on how analyze data. External reliability refers to the possibility of replication, indeed a research is externally reliable when another researcher is able to conduct the same study and obtain the same results (Saunders et al, 2016).

(33)

Several actions can be taken in order to reduce the risk of low levels of reliability. The collection of data in a standardized way such as the survey strategy ensure high levels of reliability. Indeed, questionnaires are easily replicable and usually questions are based on existing studies. Something important when a study is cross-national is ensuring that each question will be translated accurately. However, something that can reduce reliability is how people from different countries will be interpret the same question. In order to achieve high levels of external reliability it is important to ensure it also during the analysis. This is the reason why, the analysis process and the techniques used during the research will be carefully explained.

Turning to validity, it has different aspects. The first one is the measurement validity or internal validity and it measures the extent to which the data collection techniques actually measures what they want to measure (Saunders et al, 2016). Since the questionnaire used for this research entails measures used in previously made survey by ISSP that have been used for valid researches, it is possible to conclude that the questionnaire has a high internal validity. In order to ensure internal validity, the Cronbach’s Alpha analysis and the Factor Analysis have been performed.

The second aspect is external validity and it refers to the extent to which the findings can be generalized to other settings. The high number of responses should enable to have a sample that is statistically representative and that results can be generalized.

However, since the research focuses on two countries with peculiar characteristics, it is likely to believe that the research can be generalized to countries with similar characteristics.

(34)

4 Results

The analysis tries to identify differences across Norway and France on the level of job quality and organizational commitment and their relationship. These elements have been tested through regression analysis on SPSS.

The dataset for Norway and France consists of 2.774 respondents: 1.550 (55,9%) from Norway, 1.224 (44,1%) from France. The 53,8% of the respondents were females (1.493) and the 46,2% were males (1.281). The same proportions are valid when each country is selected individually.

4.1 Descriptive Statistics

Table 3 and Table 4 refers to Norway, Table 5 and 6 to France.

Gender, type of organization and the occupational categories are all dummy variables.

For gender: 1 is female, 0 is male; type of organization: 1 is public employer, 0 is private employer.

The scale of all the job characteristics and of organizational commitment have been inverted, so that a higher result means higher importance for that variable.

(35)

Table 3 - Mean and Standard Deviations in Norway

Mean SD

1 ^Age ^48,05 ^15,811

2 ^Gender ^0,54 ^0,499

3 Type of Organization 0,46 0,499

4 Managers & Professionals 0,33 0,472

5 Technicians 0,16 0,366

6 Clerks & Service Workers 0,21 0,405

7 Craft & Plant Operators 0,12 0,325

8 Job Security 3,88 1,002

9 High Income 2,87 1,016

10 Advancement Opportunities 2,60 0,945

11 Interesting Work 4,09 0,819

12 Job Autonomy 4,18 0,751

13 Organizational Commitment 3,59 0,789

Table 3 presents some descriptive statistics, precisely mean and standard deviations, of the variables included in the research model.

(36)

Table 4 displays correlation coefficients between the variables and their significance.

Table 4 - Correlations in Norway

1 2 3 4 5 6 7 8 9 10 11 12 13

1 Age 1,000

2 Gender -0,031 1,000

3 Type of

Organization ,082** ,283** 1,000

4 Managers &

Professionals ,060* 0,046 ,184** 1,000

5 Technicians 0,034 -0,040 -,077** -,308** 1,000

6

Clerks &

Service Workers

-,074** ,239** ,059* -,362** -,222** 1,000

7

Craft &

Plant

Operators 0,022 -,287** -,218** -,262** -,160** -,188** 1,000 8 Job Security -0,023 ,083** ,243** ,098** -,066* -0,008 -0,035 1,000

9 High Income -0,015 -,215** -,135** ,179** 0,037 -,213** -0,010 ,150** 1,000

10 Advancement

Opportunities -,119** -,121** -0,047 ,068* 0,038 -,084** -0,045 ,083** ,384** 1,000

11 Interesting

Work ,097** 0,043 ,151** ,242** 0,017 -,149** -,082** ,185** ,207** ,240** 1,000

12 Job

Autonomy ,098** 0,008 -0,017 ,080** 0,010 -,099** 0,015 ,062* ,146** ,091** ,355** 1,000

13 Organizational

Commitment 0,044 -0,067 -,124* 0,034 0,024 -0,073 0,039 0,065 ,301** ,208** ,490** ,225** 1,000

*. Correlation is significant at the 0.05 level (2-tailed).

**. Correlation is significant at the 0.01 level (2-tailed).

(37)

Descriptive statistics and correlations are presented also for France. Table 5 presents mean and standard deviations of the variables. Table 6 displays correlation coefficients between the variables and their significance.

Table 5 - Mean and Standard Deviations in France

Mean SD

1 ^Age ^51,58 ^16,601

2 ^Gender ^0,54 ^0,499

3 Type of Organization 0,36 0,479

4 Managers & Professionals 0,29 0,454

5 Technicians 0,21 0,404

6 Clerks & Service Workers 0,20 0,403

7 Craft & Plant Operators 0,10 0,303

8 Job Security 3,54 1,431

9 High Income 2,50 1,066

10 Advancement Opportunities 2,29 1,081

11 Interesting Work 3,96 0,935

12 Job Autonomy 3,84 1,069

13 Organizational Commitment 2,98 0,915

The ratio between female and male is the same in both countries. France has respondents with a higher age compared to Norway (51,58 vs. 48,05). Moreover, the distribution of occupations among the two countries is almost the same.

(38)

Table 6 - Correlations in France

1 2 3 4 5 6 7 8 9 10 11 12 13

1 Age 1,000

2 Gender -,121** 1,000

3 Type of

Organization ,064* ,145** 1,000

4 Managers &

Professionals ,091** -0,049 ,096** 1,000

5 Technicians -0,027 -0,033 -,061* -,324** 1,000

6

Clerks &

Service Workers

-,066* ,221** 0,023 -,323** -,257** 1,000

7

Craft &

Plant Operators

0,017 -,240** -,145** -,215** -,171** -,171** 1,000

8 Job Security 0,045 ,088* ,363** ,095* -,082* -0,017 -,133** 1,000

9 High Income 0,027 -,079* -0,022 ,264** -0,007 -,142** -,116** ,272** 1,000

10 Advancement

Opportunities -0,074 -,140** 0,057 ,132** 0,043 -,168** -0,068 ,253** ,474** 1,000

11 Interesting

Work -0,048 0,026 ,102** ,178** 0,059 -0,070 -,156** 0,050 ,174** ,274** 1,000

12 Job

Autonomy 0,006 -,077* -,161** ,082* 0,042 -,096* -0,012 -0,019 ,133** ,138** ,326** 1,000

13 Organizational

Commitment -0,020 -0,099 -0,093 -0,016 -0,043 -0,024 ,170** 0,012 ,324** ,425** ,356** ,336** 1,000

*. Correlation is significant at the 0.05 level (2-tailed).

**. Correlation is significant at the 0.01 level (2-tailed).

In both countries, age and gender are not significant correlated to organizational commitment. In Norway, being in a public organization have a slightly negative impact on organizational commitment (p ≤ .01). Both in Norway and France all the job characteristics excepts for job security are positively and significantly related to organizational commitment.

(39)

4.2 Regression Analysis

A regression analysis has been performed in SPSS Statistics in order to reject or accept the suggested hypotheses based on the theory.

In order to compare the two countries, Norway and France, a dummy variable has been created where the value 0 represents France and 1 Norway.

4.2.1 Job quality and organizational commitment

In order to evaluate whether job quality and organizational commitment are positively correlated, a regression analysis has been performed. The regression equation is:

Organizational Commitmentj = ij + ß1Ext.Factj + ß2Int.Factj + ej

Table 7 - Regression analysis - Organizational commitment and job quality Unstandardized

Coefficients Standardized

Coefficients t Sig.

B SE Beta

(Constant) 3,335 0,159 20,974 0,000

Age 0,001 0,003 0,020 0,533 0,594

Gender -0,007 0,072 -0,004 -0,096 0,924

Type of Organization -0,053 0,072 -0,029 -0,739 0,460

Managers & Professionals -0,167 0,114 -0,091 -1,467 0,143

Technicians -0,054 0,121 -0,026 -0,449 0,653

Clerks & Service Workers 0,029 0,126 0,013 0,229 0,819

Craft & Plant Operators 0,287 0,151 0,091 1,903 0,058

Extrinsic factor 0,299 0,034 0,339 8,931 0,000

Intrinsic factor 0,407 0,033 0,467 12,347 0,000

a. Dependent Variable: Organizational commitment

The R of the model is .569, while the R² is .324. This result means that 32,4% of the variance of organizational commitment is explained by the independent variables included in the model.

(40)

Hypothesis 1 stated that job quality has a positive relationship with organizational commitment. According to Table 7, both the factors composing job quality, namely extrinsic factor and intrinsic factor, are positively and significantly correlated to organizational commitment. In particular, the intrinsic factor (b = .407) has a stronger impact than the extrinsic factor (b = .299).

Job quality is analyzed in the two countries according to both factors: extrinsic and intrinsic. In the regression equation, the dummy variable of country will be used in order to evaluate the differences among the two countries.

The two regression equations are:

Extrinsic Factorj = ij + ß1Countryj + ej

Intrinsic Factorj = ij + ß1Countryj + ej

Table 8 - Regression analysis - Extrinsic factor

Unstandardized Coefficients

Standardized

Coefficients t Sig.

B SE Beta

(Constant) 0,147 0,118 1,246 0,213

Age -0,006 0,002 -0,077 -3,161 0,002

Gender -0,263 0,052 -0,130 -5,004 0,000

Type of Organization 0,240 0,052 0,118 4,633 0,000

Managers & Professionals 0,215 0,082 0,103 2,636 0,008

Technicians -0,039 0,088 -0,016 -0,446 0,656

Clerks & Service Workers -0,258 0,090 -0,102 -2,864 0,004

Craft & Plant Operators -0,329 0,105 -0,101 -3,135 0,002

Country (1=NOR) 0,335 0,050 0,164 6,707 0,000

a. Dependent Variable: Extrinsic factor

To what extent does job quality influence organizational commitment across countries?